Characterizing the 3-year course of cannabis involvement in NESARC Cannabis is the most commonly used illegal psychoactive substance in the United States and in developed nations. For instance, in the U.S., over 40 percent of 12th graders reporting a lifetime history of cannabis use. Cannabis abuse/dependence, the rate of which had increased by 18 percent in the past decade, is associated with several aspects of cognitive difficulty, comorbid substance use and psychopathology and aero-respiratory problems, making cannabis involvement a growing public health concern. Despite being a potent correlate of externalizing and internalizing disorders particularly in trajectories of alcohol, tobacco and other illegal drug use, little is know of the influences on changes in an individual's level of cannabis involvement. Why do some individuals remain occasional users while others progress to heavier and even problem use or abuse and dependence? Additionally, why do some individuals cease use while others continue, or even relapse after short-term cessation? A challenge in this area of research has been the lack of longitudinal data that is both rich in assessments of cannabis involvement and its correlates and adequately sized to allow for a dissection of the various stages of cannabis involvement. This small budget secondary data analysis R03 proposes to utilize a nationally representative U.S. sample of 43,093 adults, 34,653 of whom participated in two waves of interviews, across a 3-year follow-up period to: (a) identify groups of individuals whose cannabis involvement changes (either progression or desistence) across the 3-year follow up (i.e. movers) and groups where cannabis involvement remains stable (i.e. stayers), and (b) using predictors and correlates from socio- demographic, childhood, social/contextual, co-occurring substance use, psychopathology, negative and positive life events and treatment-related, to characterize the various transitions in cannabis involvement, ranging from non-use to occasional use (and its reverse) to change from dependence to cessation. We hypothesize that socio-demographic, early childhood and the social/contextual domains will play a more prominent role in the earlier transitions (e.g. initiation of use) while stress, psychopathology and co-occurring substance use will exert a more potent influence on later transitions (e.g. from regular use to dependence). A combination of latent variable mixture modeling (e.g. latent class and mover stayer models), survival and event history analyses and propensity score analyses will allow us to define and characterize the various changes in cannabis involvement. The results from our analyses will be valuable in designing targeted interventions, based on constellations of risk and protective influences, for a particular level of cannabis involvement. The use of existing data, on which our team publishes extensively using sophisticated statistical methodology, and the shorter project duration, provides a cost-effective approach to improving our comprehension of the process of cannabis involvement. PUBLIC HEALTH RELEVANCE: This project proposes to outline change in cannabis involvement in a nationally representative sample of U.S. adults, and to characterize the risk and protective influences that are associated with these transitions in cannabis involvement. Results from this study may be useful in designing targeted interventions for cannabis involvement.